Phosphor conversion based LED limit......

PurpleBuz said:

Are we gonna get some UV based leds instead of royal blue someday ?

would love to see some shorter wavelength Blues\UVA in my lighting spectrum without adding monos.

Click to expand...

Most of the vivid white products(COB or fixtures) use a 410nm pump to drive phosphor conversion, its pretty well established, check the Soraa light engines out as an example www.soraa.com
Cheers
Mark

soraa looks good but of course the led isn't as efficient as royal blue driven........better off adding monos.
wonder what the QER is of base uv diodes in soraa is?

crazy thing is just think what we could do if we had green leds that where as efficient at mw in mw out as royal blue we could really create a lot of orange/red photons. or amber leds even..... or even return to monos eventually lol.

well its possible and there are blurples that use 12+ different wavelengths in an attempt to 'digitize' the spectrum as you suggest

BOBBY_G said:

well its possible and there are blurples that use 12+ different wavelengths in an attempt to 'digitize' the spectrum as you suggest

Click to expand...

Yeah I understand that but the underlying problem with that approach is the inefficiency and or lack of stability of the different color diodes. Once the whole range of colors becomes quite a bit more efficient then we will see phosphor added to those as well to make broad band colors for a really nice spectrum.

If any of the higher NM diodes catch up to blue in efficiency then we will see an increase in overall efficiency of our lights.

Oh and 1 more thing..... We have all been calculating the overall efficiency and heat sink sizes wrong..... CCT will affect the overall amount of heat going into the heatsink.

Would love to have some real discussion about my theories and if they are solid......


QER of original light source(Royal Blue) is umol/j limit.

If we calculate by Royal Blue we can understand blue diode efficiency.

If we then know phosphor efficiency (actual photon loss)

We then can figure how much heat that phosphor conversion generates(the more conversion ie(warmer cct) the more heat generated.

This is how we should be figuring out total system efficiency and sizing heat sinks appropriately....



If there is a glaring problem with this please enlighten me.


Actually so it comes down to splitting hairs basically. Ie using Cree binning info isn't entirely 100% the best way to go. 4000k 80cri and 3500k 80cri in cxb3590 are in the same bin. The lumens and photon count can be identical but 4000k definitely puts out more milliwatts of energy per watt in.

Stephenj37826 said:

crazy thing is just think what we could do if we had green leds that where as efficient at mw in mw out as royal blue we could really create a lot of orange/red photons. or amber leds even..... or even return to monos eventually lol.

Click to expand...

but that's a fact of life (or should I say limitation of physical properties)

Some mfg are generating green light via phosphors instead of monos, presumably cause of efficiency.

I see where you are trying to go with this but you are lapping a few different question together that are and need to be addressed separately.

-One thing I think you're trying to say that because the excitation of phosphorus is not part of the die and it's exchange of heat through the substrate>TIM>sink, that it's conversion loss(in-efficiency) should not be calculated for sink needs? Almost like separating efficiency into thermal efficiency and then a separate radiant efficiency. It's a deep question/thought. But it isn't so easy.

-You are also asking 2 other questions,
1) what are the blue dies limits
2) what are the limits of phosphor conversion/excitation and stoke loss. (I don't know where this limit is, and use QD's are a possible avenue)

First thing to do is drop QER from you mind till the end. It means and does nothing until the end.

Blue dye have an efficiency. Efficiency is in mW, not photons. mW are constant measurements of heat...photons are variable and why we are forgetting about QER right now.

Blue die efficiency(mW out ÷ mW in) X phosphor conversion efficiency(conversion is given by most as lm/radiant watt....then ÷ that by the LER) = total white LED radiant efficiency.

Now you can factor in the QER of that specific phosphor and get a photon output. Once in radiant light is the only point at which the spectrum will affect your calculations.
The reason why we see 4k or 5k winning in both radiant and photon count over 3k is because of phosphor tech, not led die. And where the discrepancy in many people basic assumptions based on what the bin says come from.

That's what I know of it.

I tend to think of it as royal blue die @100% efficiency has x photons or 3.72 umol/j. This happens outside of phosphor conversion. I guess my question is if we start with a maximum of 3.72 umol of royal blue photons can the phosphor conversion ever yield a higher umol count than what we started with. I'm thinking no.

Stephenj37826 said:

I tend to think of it as royal blue die @100% efficiency has x photons or 3.72 umol/j. This happens outside of phosphor conversion. I guess my question is if we start with a maximum of 3.72 umol of royal blue photons can the phosphor conversion ever yield a higher umol count than what we started with. I'm thinking no.

Click to expand...

Wouldn't that carry more energy as we're converting energy (squishing waves) not actual molecules?

Stephenj37826 said:

I tend to think of it as royal blue die @100% efficiency has x photons or 3.72 umol/j. This happens outside of phosphor conversion. I guess my question is if we start with a maximum of 3.72 umol of royal blue photons can the phosphor conversion ever yield a higher umol count than what we started with. I'm thinking no.

Click to expand...

Physics say and show the opposite. Yes.
And again...why you need look take QER separately from efficiency.
You keep lapping separate questions on top of each other.
QER is just a distribution of mW just as lumens is...just shifting water to different sides of the tub...but the tub still has the same amount of water. It is all dependant on the radiant watts and a conversion...i.e. mW and LER.
Let's simplify...
100% blue(also know as 1 radiant watt output per 1watt of input) will max out at 3.72µmols/j...
100% ~660 red will produce 5.5µmol...or per radiant watt.
Both achieved with a radiant watt. That is fact and constant....you keep trying to interchange that fact, and the losses that current technology present.

If you have a 100% blue led and also 100% efficient phosphor(~325lm/radiant watt) to use over it with a post conversion QER like cree(4.7µmols/j) ...you would in fact have 4.7µmols/j.

100% blue(1 radiant watt) using a 80% conversion phosphor(260lm per radiant watt) would give off 3.76µmols/j in full spectrum form. Compared to monochromatic blue of 3.72

PS...there is no phosphor conversion, or "after the conversion" with a blue. It's just what the die is outputting.

I guess my question should be why can phosphor only produce lower energy photons than what goes in? If it's not a "peeling" of energy from photons as they pass through the phosphor layer then why can't we use any wavelength photons to excite the phosphor layer?

the equation...

with constant energie we got more µmols aka Photons on higher wavelengths


but the energie of a single Photon goes down at higher wavelengths...

https://en.m.wikipedia.org/wiki/Photoluminescence

This explained it for for me. The phosphor absorbs photon energy and creates new photons. As long as we don't create more energy we don't break the laws of physics?

Atulip said:

https://en.m.wikipedia.org/wiki/Photoluminescence

This explained it for for me. The phosphor absorbs photon energy and creates new photons. As long as we don't create more energy we don't break the laws of physics?

Click to expand...

That's kinda what I was getting at. As I still ponder can I hit the phosphor with x photons in any nm and get more photons out as in numerically? Just curiosity lol. Is it possible or is that the limitation of the phosphor itself. As much as I do understand some of this it still leaves me wanting to understand the exact reactions that take place inside the phosphor. It may be a question similar to "how many licks does it take to get to the center of a tootsie pop?" Crunch the world may never know lol.

Should be possible from what I take of it. It absorbs the photons, uses the energy at its efficiency rating, and creates photons, like how diodes create photons using electrical energy directly.

guod said:

the equation...
View attachment 3701324
with constant energie we got more µmols aka Photons on higher wavelengths
View attachment 3701331

but the energie of a single Photon goes down at higher wavelengths...
View attachment 3701332

Click to expand...

And all because the speed the light is a constant, which is pretty neat. Also, welcome to quantum mechanics.

Using some of GG's numbers here

100% blue(also know as 1 radiant watt output per 1watt of input) will max out at 3.72µmols/j..

So we have 3.72 x 6.022 x 10^17 photons

Now these photons pass through phosporus. Let's assume all photons get converted to 660nm red
The conversion is pretty efficient in a way that we don't lose photons. However high energy blue Photon get converted into lower energy 660nm photon. This is Stokes shift loss.

So now after phosporus I get same number of photons
3.72 x 6.022 x 10^17 photons

But 1w 100% 660nm should have given me
5.5 x 6.022 x 10^17 photons

(5.5-3.72)/5.5= I got same photons after phosporus but lost 32.36% energy as heat in phosporus Stokes shift loss

So even with 100% efficiency of blue we cannot exceed the number of initial photons created.

Edit:
We do Lm/w and divide by LER to get efficiency.
Then we multiply QER of that spectrum with efficiency to get u moles/j. This u mol/j can never exceed the u mol/j of base blue diode.

Assuming UV and Blue have same efficiency, This also means using UV instead of blue is a bad idea because UV more energy and we will end up losing more in Stokes shift loss.

@robincnn I thought we were absorbing photons, then using the energy to create new photons. The only limit should be the amount of energy. Or does phosphor operate differently than photoluminescence?

https://en.m.wikipedia.org/wiki/Photoluminescence

robincnn said:

We do Lm/w and divide by LER to get efficiency.
Then we multiply QER of that spectrum with efficiency to get u moles/j. This u mol/j can never exceed the u mol/j of base blue diode.

Click to expand...

what is the QER of a white Cob?

Now somebody explain to me how a 90CRI white led can be more photon efficient than an 80CRI white led at the same CCT?